Members of cdc2 family of protein kinases are known for their pivotal role in the regulation of the eukaryotic cell cycle. The potential roles of neuronal specific cdc2-like kinases in stabilizing neurofilament skeleton and in axonal morphogenesis through tau by phosphorylation are not well delineated. The main objectives of this project are: (1) to disrupt the genomic locus of neuronal cdc2-like kinase (cdk5) in embryonic stem cells, and then use the targeted cells to generate mouse models to study in vivo function of these kinases; and (2) to overexpress cdk5 in neuronal cell lines and in vivo in mice. Toward achieving these objectives, we have isolated 8 genomic clones of cdk5 from the 129/svj library. These isogeneic clones have been partially characterized to define intron/exon boundaries and exonic sequences. A gene-targeting construct has been engineered with a deletion of exons 3 through 5 and insertion of neomycin resistance gene at the site of deletion. At the 3' end of the construct, the herpes virus thymidine kinase gene is included for positive/negative selection using G418 and gancyclovir. We have also isolated a murine cDNA clone using the RT-PCR technique, and have obtained complete nucleic acid sequence data. Gene targeting experiments in ES are currently in progress to obtain targeted clones.